Testing an artificial intelligence tool to reduce the spread of airborne infections in hospitals
| ISRCTN | ISRCTN12421486 |
|---|---|
| DOI | https://doi.org/10.1186/ISRCTN12421486 |
| Integrated Research Application System (IRAS) | 353916 |
| Central Portfolio Management System (CPMS) | 67415 |
| National Institute for Health and Care Research (NIHR) | 205439 |
| Sponsor | University College London |
| Funder | National Institute for Health and Care Research |
- Submission date
- 23/10/2025
- Registration date
- 19/11/2025
- Last edited
- 19/11/2025
- Recruitment status
- Not yet recruiting
- Overall study status
- Ongoing
- Condition category
- Infections and Infestations
Plain English summary of protocol
Background and study aims
The Air Safety programme research team has developed an Artificial Intelligence air Safety Tool (AISaT): a computer software that guides users on how to best reduce airborne infection risks in hospitals, using cheap solutions like air filters, fans and screens. Clinical trials are a key part of the programme to assess the effectiveness of AISaT recommendations and to demonstrate reduced infection risks. This study aims to assess the effectiveness of AISaT recommendations in outpatient, day-case rooms, and wards to reduce risks of airborne disease transmission.
Who can participate?
Clinicians working at relevant clinical testing environments: outpatient clinic rooms, Aerosol Generating Procedure (AGP) rooms and hospital ward bays. All patients (and accompanying persons) attending those relevant clinical testing environments.
What does the study involve?
A small aerosol-producing device(s) will be installed within the room to generate saline aerosol droplets, and air particle counters (APCs) will be installed next to participating clinicians to measure the primary outcome, which is the average number of aerosol droplets per minute per clinical encounter as measured by an air particle counter. There will also be questionnaires and interviews.
What are the possible benefits and risks of participating?
Benefits:
Participants may benefit from improved air quality in clinical spaces due to the implementation of AISaT mitigation strategies. These strategies aim to reduce airborne particles and improve ventilation, potentially lowering the risk of exposure to airborne infections.
Risks:
The known and potential risks include trip hazards in all study stages when air safety mitigation devices are placed in unsuitable places when following the AISaT guidance, for example, in the middle of the floor, with leading wires left uncovered.
Where is the study run from?
University College London, UK.
When is the study starting and how long is it expected to run for?
January 2026 to September 2027
Who is funding the study?
National Institute for Health and Care Research (NIHR), UK.
Who is the main contact?
Laurence Lovat, l.lovat@ucl.ac.uk
Contact information
Public, Scientific, Principal investigator
Charles Bell House
43–45 Foley Street
London
W1W 7TS
United Kingdom
 ORCID ID |
0000-0003-4542-3915 |
|---|---|
| Phone | +44207679 9083 |
| l.lovat@ucl.ac.uk |
Study information
| Primary study design | Interventional | |
|---|---|---|
| Allocation | Randomized controlled trial | |
| Masking | Blinded (masking used) | |
| Control | Placebo | |
| Assignment | Crossover | |
| Purpose | Health services research, Prevention, Screening | |
| Participant information sheet | 48268_PIS_17Jul2025.pdf | |
| Scientific title | Assessing effectiveness of Artificial Intelligence air Safety Tool (AISaT) recommendations in outpatient, day-case rooms, and wards to reduce risks of airborne disease transmission | |
| Study acronym | AISaT | |
| Study objectives | These trials are to investigate: A) Whether placing mitigations for airborne disease transmission according to the recommendations of the AISaT software tool reduces airborne disease transmission risks in outpatient consulting rooms, aerosol-generating procedure rooms and hospital wards. B) Whether it is feasible to implement these mitigations at different hospital sites (and identify any variation in implementation). C) Whether patients and hospital staff find the AISaT recommendations acceptable Primary Objective: To determine in hospital consulting rooms, aerosol generating procedure rooms and wards, when using mitigations for airborne disease transmission according to recommendations of the AISaT software tool, the reduction in the relative number of airborne aerosol droplets per standard time period breathed by any staff member present compared to when no mitigations are used or when mitigations are used as decided by the clinician using the relevant space. Outcome Measure: Number of aerosol droplets on average per minute per clinical encounter as measured by an air particle counter (APC) installed next to the clinician and at up to 4 other fixed locations in the room or ward bay. Secondary Objective: Mixed methods evaluation of staff experiences of using the tool and the mitigation approaches that are recommended (based on analysis of acceptability by staff and feasibility of implementation). | |
| Ethics approval(s) |
Approved 05/08/2025, East Midlands – Leicester Central Research Ethics Committee (2 Redman Place, Stratford, London, E20 1JQ, United Kingdom; +44 (0)207 104 8118; leicestercentral.rec@hra.nhs.uk), ref: 25/EM/0156 | |
| Health condition(s) or problem(s) studied | Airborne disease transmission in hospitals | |
| Intervention | The Air Safety programme research team has developed an Artificial Intelligence Air Safety Tool (AISaT) - a computer software that guides users on how to reduce airborne infection risks in hospitals using cost-effective solutions like air filters, fans, and screens. The research programme will investigate whether AISaT works, is acceptable, and cost-effective. Two clinical trials will be run in two hospitals: University College London Hospital (UCLH), London Lister Hospital, Stevenage (East and North Hertfordshire NHS Trust) Study Process in Chronological Order: Initially, clinicians at both sites will be identified, screened and invited to participate in the study. Obstetric, psychiatric and paediatric clinical areas will be excluded. If they consent, they will then be asked some demographic questions as stipulated in the protocol (activity level of participants, sex, age to nearest decade, approximate height and weight, ethnicity, history of respiratory difficulties). Appropriate weekly clinics run by the participant clinician will then be identified as the dates that the study team will attend and run the study. This will take place over 3 consecutive weeks. Clinicians will be randomly assigned to one of three room configurations: Configuration 1 - AISaT will be turned on, and air safety measures (air filters, fans, screens) will be implemented. Configuration 2 - AISaT will be turned off completely with no extra air quality safety measures. Configuration 3 - Air quality safety measures will be offered to the clinician to use and place as they see fit. Patients due to have an appointment with an enrolled clinician on the clinic day will be screened (the only exclusion criterion is that they cannot be under 18 years old). Information leaflets will be sent to all eligible patients attending relevant clinics by post or email at least 24 hours before they attend for their appointments. Potential participants will then be approached by a member of the research team on the day that they attend to sign a consent form. If patients have an accompanying person attending their appointment with them, they will also need to be consented before the appointment. After consenting, the patient and accompanying persons will also complete the demographic questions. Patients and accompanying persons will not be randomised. The consultation will then take place exactly the same as it would have as normal. The only difference is that the room will have been set up with some equipment in it depending on which configuration the clinician was assigned to. This includes air particle counters next to the clinician and up to 4 other fixed locations in the room, a device used for the Al software 'AISaT' (laptop or tablet), a small aerosol-producing device (emitting small aerolised droplets of saline, which is salt water), air filter(s) and an infrared camera to generate heat maps of the room. The air particle counters will measure the primary outcome, which is the average number of aerosol droplets per minute per clinical encounter. The clinical trials will assess AISaT's effectiveness following the above process in three settings: 1. Outpatient consulting rooms 2. Aerosol-generating procedure rooms (such as endoscopy suites) 3. Wards Alongside the clinical trials, qualitative implementation research will also be conducted, which will involve exploring issues such as usability and acceptability of the AISaT tool with key stakeholders (such as patients, clinicians and estates staff). The research will explore how a successful AISaT tool would be implemented and identify barriers to implementation. A subset of these participants will be invited to take part in the implementation research (e.g. clinicians, patients and estates staff). There are separate optional consenting processes for each of the following: ethnography research (staff only), usability testing (staff and patients), questionnaires and interviews (staff and patients). Please see the uploaded topic guides for interviews, as well as the Normalisation MeAsure Development (NoMAD) questionnaire. Any personal data collected as part of the qualitative evaluation will be stored in a data safe haven, a technical solution for storing, handling and analysing identifiable data. The UCL Data Safe Haven has been certified to the ISO27001 information security standard and conforms to NHS Digital's Information Governance Toolkit. For the Air Safety NIHR programme, it has been contractually agreed that UCL will own IP. Data will be transferred from UCLH to UCL, as well as from East and North Hertfordshire NHS Trust to UCL. Personalised data will be identifiable and stored in UCL Data Safe Haven. The data controller for this project will be University College London (UCL). The UCL Data Protection Officer provides oversight of UCL activities involving the processing of personal data, and can be contacted at data-protection@ucl.ac.uk. The data processor is University College London and East and North Hertfordshire NHS Trust (for Lister Hospital). The data custodian is Professor Ramani Moonesinghe, Professor of Perioperative Medicine at UCL and Chief Investigator of the Central London Patient Safety Research Collaborative. | |
| Intervention type | Drug/Device | |
| Phase | Not Applicable | |
| Drug / device / biological / vaccine name(s) | Artificial Intelligence Air Safety Tool (AISaT) | |
| Primary outcome measure(s) |
Number of aerosol droplets on average per minute per clinical encounter measured using an air particle counter (APC) installed next to the clinician and at up to 4 other fixed locations in the room or ward bay at each of the sessions over the 3-week intervention period | |
| Key secondary outcome measure(s) |
1. Implementation processes are measured using the Normalisation Measure Development Questionnaire (NoMAD), collected at one time point alongside the clinical trial | |
| Completion date | 27/09/2027 |
Eligibility
| Participant type(s) | Health professional, Patient |
|---|---|
| Age group | Mixed |
| Lower age limit | 18 Years |
| Upper age limit | 120 Years |
| Sex | All |
| Target sample size at registration | 4848 |
| Key inclusion criteria | 1. Clinicians working at relevant clinical testing environments (1st trial outpatient clinic room, 2nd trial AGP room, 3rd trial ward bay) 2. All patients (and accompanying persons) attending those relevant clinical testing environments 3. The mixed method evaluation may include people working in estates teams, hospital managers and other healthcare staff. |
| Key exclusion criteria | 1. Obstetric, psychiatric and paediatric clinical areas will be excluded, to minimise the risks of ethical complications involving children, pregnant women or patients with psychiatric illnesses that may have difficulty consenting 2. People under the age of 18 years old |
| Date of first enrolment | 05/01/2026 |
| Date of final enrolment | 27/07/2027 |
Locations
Countries of recruitment
- United Kingdom
- England
Study participating centres
London
NW1 2PG
England
Coreys Mill Lane
Stevenage
SG1 4AB
England
Results and Publications
| Individual participant data (IPD) Intention to share | Yes |
|---|---|
| IPD sharing plan summary | Available on request |
| IPD sharing plan | The datasets generated during and/or analysed during the current study will be available upon request from Chief Investigator Professor Laurence Lovat, email: l.lovat@ucl.ac.uk. 1. The type of data that will be shared The study will be collecting the following personal data, which will be collected directly from participants and from the medical records of patient participants: activity level of participants, sex, age, height, weight, ethnicity, and history of respiratory difficulties. The Case Report Forms (CRFs) will also include the participant’s initials and trial identification number. For all subsequent analyses, pseudo-anonymised data will be used. 2. Timing for availability Data will be stored for the time recommended by UCL for non-interventional clinical trials used in regulatory submissions, as it is conceivable that a regulatory submission could develop following this work. This is set at 10 years. 3. Whether consent from participants was required and obtained This will be clearly explained in the participant information sheet (PIS). Participant consent for this will be sought. 4. Comments on data anonymization For all subsequent analyses, pseudo-anonymised data will be used. The data and the linking code will be maintained securely in separate locations using encrypted digital files within password-protected folders and storage media. Identifiable data will not be transferred outside the study team or outside UCL. 5. Any ethical or legal restrictions The study will be compliant with the requirements of the General Data Protection Regulation (2016/679) and the UK Data Protection Act (2018). All Investigators and study site staff will comply with the requirements of the General Data Protection Regulation (2016/679) with regard to the collection, storage, processing and disclosure of personal information, and will uphold the Act’s core principles. 6. Any additional comments UCL is the data controller; the UCL Data Protection Officer is data-protection@ucl.ac.uk. The data processor is University College London and East and North Hertfordshire NHS Trust (for Lister Hospital). The data custodian is Professor Ramani Moonesinghe, Professor of Perioperative Medicine at UCL and Chief Investigator of the Central London Patient Safety Research Collaborative. Data access will be limited to the minimum number of individuals necessary for quality control, audit and analysis. Participants will have provided informed consent for their anonymised data to be used in future ethically approved research. All shared data will be fully de-identified in accordance with GDPR and UCL data protection policies. Access will be granted only to researchers with appropriate ethical approval and data sharing agreements in place. There are no known legal or ethical restrictions beyond those described above. |
Study outputs
| Output type | Details | Date created | Date added | Peer reviewed? | Patient-facing? |
|---|---|---|---|---|---|
| Participant information sheet | 17/07/2025 | 05/11/2025 | No | Yes | |
| Participant information sheet | Participant information sheet | 11/11/2025 | 11/11/2025 | No | Yes |
| Protocol file | 17/07/2025 | 05/11/2025 | No | No | |
| Study website | Study website | 11/11/2025 | 11/11/2025 | No | Yes |
Additional files
- 48268_Protocol_v.1.1_17Jul2025.pdf
- Protocol file
- 48268_PIS_17Jul2025.pdf
- Participant information sheet
Editorial Notes
23/10/2025: Study's existence confirmed by Health Research Authority (HRA) (UK)
